/*

Ryan Enos
Proposal title: Is Modern Racism Caused by Anti-Black Affect? An Experimental 
Investigation of the Attitudes Measured by Modern Racism Scales

HYPOTHESES

Stated-Hyp 1: People will express similar levels of resentments towards Black vs non-Black target groups.

	Test-Hyp 1: Participants in Black target condition will report the same amount of resentment (mean) as participants in the Surinamese or Mauritanian target condition.
	
Stated-Hyp 2: Conservatives will express similar levels of resentment across target groups.

	Test-Hyp 2: Conservative participants in Black target condition will report the same amount of resentment (mean) as conservative participants in the Surinamese or Mauritanian target condition.

	
Stated-Hyp3: Liberals will express less resentment towards Blacks than other groups.

	Test-Hyp 3: Liberal participants in Black target condition will report less resentment (mean) than liberal participants in the Surinamese or Mauritanian target condition.

********************************************************************************

NOTES: 
	
There were hypotheses about median tests in the proposal. We decided to not 
conduct those tests.
	
*/

clear all
use "Enos700.dta", clear

********************************************************************************

* RECODING

* experiment condition
	tab xtess196, mis
	tab xtess196, nolabel
	recode xtess196 (1=1) (nonmiss=0), gen(treat_black)
	lab def treat_black 0 "0 Non-Black" 1 "1 Black"
	lab val treat_black treat_black 
	tab treat_black xtess196, mis

* DV
	
	tab a1a, mis
	recode a1a (-1 = .)
	tab a1a, mis

	tab a1b, mis
	recode a1b (-1 = .)
	tab a1b, mis

	tab a1c, mis
	recode a1c (-1 = .)
	tab a1c, mis

	tab a2a, mis
	recode a2a (-1 = .)
	tab a2a, mis

	tab a2b, mis
	recode a2b (-1 = .)
	tab a2b, mis

	tab a2c, mis
	recode a2c (-1 = .)
	tab a2c, mis

	tab a3a, mis
	recode a3a (-1 = .)
	tab a3a, mis

	tab a3b, mis
	recode a3b (-1 = .)
	tab a3b, mis

	tab a3c, mis
	recode a3c (-1 = .)
	tab a3c, mis
		
	gen resentment = .
	replace resentment = (a1a + a1b + a1c) if xtess196 == 1
	replace resentment = (a2a + a2b + a2c) if xtess196 == 2
	replace resentment = (a3a + a3b + a3c) if xtess196 == 3
	tab resentment, mis

* Ideology

	tab ideo_new, mis
	recode ideo_new (-1 = .)
	tab ideo_new, mis

********************************************************************************

* ANALYSIS
	

* Test-Hyp 1: Participants in Black target condition will report the same amount of resentment (mean) as participants in the Surinamese or Mauritanian target condition.

	reg resentment i.treat_black
	// no support for H1 (p = .020)
	tess 1.treat_black, init(Enos700) 

	
* Test-Hyp 2: Conservative participants in Black target condition will report the same amount of resentment (mean) as conservative participants in the Surinamese or Mauritanian target condition.

	reg resentment i.treat_black if ideo_new > 4
	// support for H3 (p = .427)
	tess 1.treat_black
	

	
* Test-Hyp 3: Liberal participants in Black target condition will report less resentment (mean) than liberal participants in the Surinamese or Mauritanian target condition.

	reg resentment i.treat_black if ideo_new < 4
	// no support for H5 (p = .051)
	tess 1.treat_black -

